1. Field of the Invention
The invention relates in general to a method for monitoring the performance of a computer network, and in particular to a method for monitoring and controlling hardware devices of a network, such as for example switches, routers and firewalls. The invention further relates to an appropriate computer network system, a computer program and a computer program product.
2. Background Art
Related technology is discussed in U.S. Pat. Nos. 6,108,782, 6,708,171, 5,109,486 and 4,817,080 and incorporated herein by reference.
Furthermore, familiarity with general concepts, protocols and devices currently substantially used in LAN (local area network) networking applications and in WAN (wide area network) internetworking applications is presumed. These standards are publicly available and not discussed herein in more detail. This specification also presumes some familiarity with specific network and operating system components discussed briefly in the following paragraph, such as the simple network management protocol (SNMP) for management of LAN and WAN networks, and the general functionality of a network proxy.
A local area network (LAN) is an arrangement of various hardware and software elements that operate together to allow a number of digital devices to exchange data within the LAN and also may include internet connections to external wide area networks (WANs). Typical modern LANs are comprised of one to many LAN intermediate systems that are responsible for data transmission throughout the LAN and a number of end systems that represent an end user equipment. End systems may be familiar end-user data processing equipment such as personal computers, workstations, and printers and additionally may be digital devices such as digital telephones or real-time video displays. Different types of end systems can operate together on the same LAN.
The already mentioned simple network management protocol (SNMP) is a common protocol used for managing network infrastructure over the network. SNMP is a layer 7 network and system management protocol that handles network and system management functions and can be implemented as a driver or SNMP agent interfacing through UDP or some other layer 4 protocol. Prior art SNMP installations largely were not placed in end systems because SNMP did not handle end system management or monitoring functions and because SNMP agents are processor and memory intensive.
SNMP is designed to provide a simple but powerful cross platform protocol for communicating complex data structures important to network infrastructure management. However, its power and platform-independent design makes it computationally intensive to implement, and for that reason it has limited applications in end system management or monitoring. It is primarily used in network infrastructure management, such as management of network routers and bridges.
In conventional computer performance monitoring applications or network management tools, the network management tool generates a request for data which is sent by the network management tool to a hardware device of the network which has to be controlled. The hardware device sends in return the information requested to the network management tool which then processes the data as required.
The hardware devices to be monitored are often of high complexity. The information to be monitored becomes continuously more complicated so that there are enormous amounts of information to be analyzed. Furthermore, a lot of different network management tools exist to monitor and control networks or network hardware devices, respectively. Such network management tools are, for example, fault management, configuration management, accounting, that means for example bandwidth measurements, performance, such as memory usage or CPU usage, security management, as, for example, firewall management, authentication, and userID management. These network management tools cause a high network load and impose more and more load on a single hardware device which can result in performance degradation of both network hardware devices as well as in network bandwidth. Each operator often has his own solution for monitoring and does not in general take into account other possibilities. Since a lot of users, particularly those who work worldwide, use a plurality of network components of different operators, there are several different network management tools which cause redundancy by monitoring hardware devices.
Different users, which have their own internal network, often use identical IP address ranges. By taking over those users as outsourcing users, address conflicts result. At the moment, the most common solution is to install a double monitoring system for each user, which results in double costs.
There exist so-called NAT (Network Address Translation) devices/tools that perform an address translation into identical IP addresses. These tools change only the IP address in the so-called TCP/IP header, but not the IP address inside the transported SNMP packet. The SNMP data received therefore still includes the duplicate IP addresses. Most network management tools cannot handle these duplicate addresses and therefore dedicated installations must be chosen instead of cheaper shared solutions.